Noninvasive Detection of Appliance Utilization Patterns in Residential Electricity Demand

Smart meters with automatic meter reading functionalities are becoming popular across the world. As a result, load measurements at various sampling frequencies now available. Several methods have been proposed to infer device usage characteristics from household measurements. However, many techniques based on highly intensive computations that incur heavy computational costs; moreover, they often rely private information. In this paper, we propose technique for detection of appliance utilization patterns using low-computational-cost algorithms do not require any information about households. Appliance identified only system status behavior, represented by large datasets, dimensionality reduction and clustering algorithms. Principal component analysis, k-means, elbow method used define clusters, minimum spanning tree is visualize results show appearance patterns. Self organizing maps create classifier. We applied our two public datasets different countries, United Kingdom (UK-DALE) US (REDD), The enable effective demand-side management, while classifier can detect malfunctions through analyses.